Efficient Optimization Algorithms for Nonlinear Data Analysis

Identification of low-dimensional structures and main sources of variation from multivariate data are fundamental tasks in data analysis. Many methods aimed at these tasks involve solution of an optimization problem. Thus, the objective of this thesis is to develop computationally efficient and theoretically justified methods for solving such problems. Most of the thesis is based on a statistical model, where ridges of the density estimated from the data are considered as relevant features. Finding ridges, that are generalized maxima, necessitates development of advanced optimization methods. An efficient and convergent trust region Newton method for projecting a point onto a ridge of the underlying density is developed for this purpose. The method is utilized in a differential equation-based approach for tracing ridges and computing projection coordinates along them. The density estimation is done nonparametrically by using Gaussian kernels. This allows application of ridge-based methods with only mild assumptions on the underlying structure of the data. The statistical model and the ridge finding methods are adapted to two different applications. The first one is extraction of curvilinear structures from noisy data mixed with background clutter. The second one is a novel nonlinear generalization of principal component analysis (PCA) and its extension to time series data. The methods have a wide range of potential applications, where most of the earlier approaches are inadequate. Examples include identification of faults from seismic data and identification of filaments from cosmological data. Applicability of the nonlinear PCA to climate analysis and reconstruction of periodic patterns from noisy time series data are also demonstrated. Other contributions of the thesis include development of an efficient semidefinite optimization method for embedding graphs into the Euclidean space. The method produces structure-preserving embeddings that maximize interpoint distances. It is primarily developed for dimensionality reduction, but has also potential applications in graph theory and various areas of physics, chemistry and engineering. Asymptotic behaviour of ridges and maxima of Gaussian kernel densities is also investigated when the kernel bandwidth approaches infinity. The results are applied to the nonlinear PCA and to finding significant maxima of such densities, which is a typical problem in visual object tracking.

A cell spot microarray method for high-throughput biology

High-throughput screening of cellular effects of RNA interference (RNAi) libraries is now being increasingly applied to explore the role of genes in specific cell biological processes and disease states. However, the technology is still limited to specialty laboratories, due to the requirements for robotic infrastructure, access to expensive reagent libraries, expertise in high-throughput screening assay development, standardization, data analysis and applications. In the future, alternative screening platforms will be required to expand functional large-scale experiments to include more RNAi constructs, allow combinatorial loss-of-function analyses (e.g. genegene or gene-drug interaction), gain-of-function screens, multi-parametric phenotypic readouts or comparative analysis of many different cell types. Such comprehensive perturbation of gene networks in cells will require a major increase in the flexibility of the screening platforms, throughput and reduction of costs. As an alternative for the conventional multi-well based high-throughput screening -platforms, here the development of a novel cell spot microarray method for production of high density siRNA reverse transfection arrays is described. The cell spot microarray platform is distinguished from the majority of other transfection cell microarray techniques by the spatially confined array layout that allow highly parallel screening of large-scale RNAi reagent libraries with assays otherwise difficult or not applicable to high-throughput screening. This study depicts the development of the cell spot microarray method along with biological application examples of high-content immunofluorescence and phenotype based cancer cell biological analyses focusing on the regulation of prostate cancer cell growth, maintenance of genomic integrity in breast cancer cells, and functional analysis of integrin protein-protein interactions in situ.

Innovaatio-ohjelmatoiminnan suorituskyvyn mittaus

Tutkimuksen tavoitteena on luoda viitekehys yritysten innovaatiotoimintaa ohjelmatoiminnalla tukevan julkisen organisaation suorituskyvyn mittaamiseksi. Tutkimus on luonteeltaan tapaustutkimus ja tutkimusotteena siinä käytetään toiminta-analyyttistä tutkimusotetta. Tutkimuksen empiirinen aineisto kerättiin havainnoimalla, keskusteluilla, haastatteluilla sekä ryhmätyöskentely-menetelmillä. Luodun viitekehyksen mukaan innovaatiotoimintaa ohjelmatoiminnalla tukevan julkisen organisaation tuottama julkinen arvo syntyy pienentämällä innovaatio-toiminnan markkina- ja järjestelmähäiriöitä, joka tapahtuu tarjoamalla rahalista tukea sekä palveluita. Näin innovaatiotoiminnan tuloksellisuus ja positiiviset ulkoisvaikutukset lisääntyvät. Tämä johtaa lopulta talouden kasvamiseen sekä yhteiskunnan ja ympäristön hyvinvointiin. Innovaatiotuella voidaan säädellä myös innovaatiotoiminnan aluetta ja näin luoda arvoa pureutumalla yhteiskunnan ongelmakohtiin. Tämän lisäksi kansalaisille on todennäköisesti arvokasta organisaatiota kohtaan tuntema luottamus, jota kasvattavat luultavasti esimerkiksi resurssien tehokas käyttö, oikeudenmukainen toiminta, asiakkaiden yhden-mukainen kohtelu, todennetut tulokset ja vaikutukset sekä toiminnan läpinäkyvyys. Viitekehyksen pohjalta muodostettiin Ohjelmat-ydinprosessin suorituskyvyn mittaristo. Aluksi kuvattiin ydinprosessin tavoitteet, joita tarkennettiin vaikuttavuusmallilla sekä sidosryhmäanalyysillä. Seuraavaksi muodostettiin ydinprosessille strategiakartta ja tunnistettiin sen kriittiset menestystekijät. Tuloksena syntyi mittaristo, jossa on 84 potentiaalista prosessin suorituskykyä mittaavaa mittaria. Lopuksi mittaristoa tarkasteltiin kokonaisuutena ja todettiin se kohtalaisen hyvin onnistuneeksi.